結合類神經網路及Kinect深度攝影機之跌倒偵測系統

簡易檢索 / 詳目顯示

回結果列表

研究生：	廖家偉 Jia-Wei Liao
論文名稱：	結合類神經網路及Kinect深度攝影機之跌倒偵測系統 A Fall Detect System Based on Neural Networks with Kinect Depth-Camera
指導教授：	蘇木春 Mu-Chun Su
口試委員:
學位類別：	碩士 Master
系所名稱：	資訊電機學院 - 資訊工程學系 Department of Computer Science & Information Engineering
論文出版年：	2016
畢業學年度：	104
語文別：	中文
論文頁數：	107
中文關鍵詞：	Kinect 、跌倒偵測、類神經網路、影像監控
外文關鍵詞：	video surveilleance, neural netwroks
相關次數：	點閱：11 下載：0
分享至:	分享至facebook 分享至twitter

查詢本校圖書館目錄查詢臺灣博碩士論文知識加值系統勘誤回報

近年來社會面臨老年化日趨嚴重的問題，老人照護議題也越發重要；老
人跌倒的發生頻率高，且常附帶著因延後就醫所帶來的巨大風險，因此跌倒
偵測的相關研究越來越蓬勃發展。本論文開發一套基於結合Kinect 及類神
經網路的跌倒偵測系統，希望在多人情境下，能夠正常運作，使其更貼近現
實生活。本系統使用Kinect 所提供之深度資訊，找出場景中的地面資訊，
再搭配背景相減演算法，找出前景資訊，並進一步追蹤；再利用事前訓練好
的類神經網路模型及選定的特徵，判斷是否有跌倒事件發生，當系統偵測跌
倒時，會記錄當下的畫面以及時間，回傳給照顧者。本論文會針對規則式判
斷可能會發生誤判的情況做探討，及利用類神經網路所帶來的優勢。
本論文設計了六種模擬情境，三種單人情境、三種多人情境，並且比較
規則式判斷及使用類神經網路的實驗結果，探討之間的差別以及誤判的可
能因素。在全部的情境中總共有168 次跌倒事件以及168 次未跌倒事件，
其實驗結果，敏感度(Sensitivity)約為97%，特異度(Specificity)約為90%，
Kappa 值為0.84，證明系統有幾乎與事實吻合的程度。

Recently, society is faced with the problematic issue of an aging population.
The eldercare issue is extremely important. The frequency of falls in the elderly
is higher than in younger people with a greater risk caused by treatment delay.
Therefore, the research of fall detection systems has been increasing drastically.
This thesis proposes to develop a fall detection system based on neural networks
with Kinect depth-camera. We hope it can operate reliable in a complex
environment or in multi-person scenarios. The system uses raw data of Kinect
depth images to locate the ground in the scene, identify the foreground pixels with
a background subtraction algorithm, and then tracked the foreground for analysis.
Last, the system will judge whether the fall events occurred by using its welltrained
neural networks model and the specified features. When fall events are
detected, the system would record the image and time immediately, then report to
caregivers for efficient aid. Additionally, this thesis will discuss the reasons for
rule decision system’s misjudgment and the advantages of using neural networks.
The performance of the proposed system was verified by six experimental
scenarios. There are three single person and for multi-person experimental
scenarios. After these experiments, we would compare the result of rule decision
system with the proposed system and discuss the difference and the reason of
misjudgment between both of them. Among all of these experimental scenarios:
168 are fall events and 168 are not fall events. The results show the sensitivity
iii
rate and the specificity rate were 97% and 90%, respectively. And the Kappa value
of the proposed system is 0.84 which is higher than 0.80, showing that we have a
reliable system that accurately reflects reality in terms of fall events.

摘要  .......................................................................i
ABSTRACT ...................................................................ii
誌謝 .......................................................................iv
目錄 ........................................................................v
圖目錄 ....................................................................vii
表目錄 .....................................................................ix
第一章、 緒論 .............................................................. 1
1-1 研究動機 研究動機 ...................................................... 1
1-2 研究目的 研究目的 ...................................................... 3
1-3 論文架構 論文架構 ...................................................... 4
第二章、 相關研究 .......................................................... 5
2-1 跌倒偵測 跌倒偵測 ...................................................... 5
2-1-1 感測 ..................................................................5
2-1-2 影像式偵測系統 ....................................................... 7
2-1-3 加入機器學習 ......................................................... 9
2-2 類神經網路(Neural Networks)  .......................................... 11
2-2-1 類神經網路簡介 ...................................................... 11
2-2-2 倒傳遞類神經網路 演算法 (Back-propagation networks) ..................12
2-2-3 放射狀基底函數類神經網路  (Radial Basis Function Networks, RBFN) .... 19
2-2-4 支撐向量機 (Support Vectors Machines, SVM) .......................... 24
2-2-5 深度學習 (Deep Learning) .............................................29
第三章、 跌倒偵測系統 ..................................................... 34
3-1 系統架構 系統架構 ..................................................... 35
3-1-1 硬體介紹 ............................................................ 35
3-1-2 軟體架構 ............................................................ 36
3-1-3 系統介面說明 ........................................................ 37
3-2 環境建置 環境建置 ..................................................... 39
3-2-1 座標轉換 ............................................................ 39
3-2-2 背景相減法 ...................................................... ... 44
3-3 簡單規則判斷式的跌倒偵測系統 .............................. ........... 46
3-4 特徵擷取 特徵擷取 ..................................................... 48
3-5 類神經網路訓練 類神經網路訓練 ......................................... 52
第四章、 實驗設計與結果 ................................................... 55
4-1 實驗設計 實驗設計 ..................................................... 55
4-1-1 評估指標 ............................................................ 55
4-1-2 情境設計 ............................................................ 58
4-1-3 不同之類神經網路比較  ............................................... 62
4-1-4 K 折交叉驗證法 (K -Fold cross validation) [40] ...................... 63
4-2 實驗結果 實驗結果 ..................................................... 64
4-2-1 規則式判斷跌倒偵測實驗結果  ......................................... 64
4-2-2 四種分類器以及 Neuroph Studio 訓練測試結果 .......................... 66
4-2-3 醫院模擬情境實驗結果  ............................................... 76
4-2-4 卷積類神經網路實驗結果 .............................................. 82
4-2-5 類神經網路跌倒偵測實驗結果  ......................................... 85
第五章、 結論與 未來展望 .................................................. 87
5-1 結論 .................................................................. 87
5-2 未來展望  ............................................................. 89
參考文獻 .................................................................. 90
                                

[1]聯合報:逾四成老人在家中跌倒，浴室、樓梯最危險.[Online]. Available: http://www.uho.com.tw/hotnews.asp?aid=39414. [Accessed: 09-Jun-2016].
[2] 台灣病人通報資訊網. [Online]. Available:
http://117.56.33.147/Content/Downloads/List01.aspx?SiteID=1&MmmID=621273303702500244. [Accessed: 09-Jun-2016].
[3] J. Y. Hwang, J. M. Kang, Y. W. Jang, and H. C. Kim, “Development of Novel Algorithm and Real-time Monitoring Ambulatory System Using Bluetooth Module for Fall Detection in the Elderly,” in 26th Annual International Conference of the IEEEEngineering in Medicine and Biology Society, pp. 2204-2207, 2004.
[4] M. Alwan, P. J. Rajendran, S. Kell, D. Mack, S. Dalal, M. Wolfe, and R. Felder, “A Smart and Passive Floor-Vibration Based Fall Detector for Elderly,” in 2nd International Conference on Information andCommunication Technologies, pp. 1003-1007, 2006.
[5] Y. Lee, J. Kim, M. Son, and M. Lee, “Implementation of Accelerometer Sensor Module and Fall Detection Monitoring System based on Wireless Sensor Network,” in 29th Annual International Conference ofthe IEEE Engineering in Medicine and Biology Society, pp. 2315-2318, 2007.
[6]F. Sposaro and G. Tyson, “iFall: An android application for fall monitoring and response,” in 2009 Annual International Conference of

the IEEE Engineering in Medicine and Biology Society, pp. 6119-6122, 2009.
[7] A. Sixsmith and N. Johnson, “A smart sensor to detect the falls of the elderly,”inIEEE Pervasive Computing, vol. 3, no. 2, pp. 42-47, 2004.
[8] D. Anderson, J. M. Keller, M. Skubic, X. Chen,and Z. He, “Recognizing Falls from Silhouettes,”in 28th Annual International Conference of the IEEEEngineering in Medicine and Biology Society, pp. 6388-6391, 2006.
[9] C. Rougier, J. Meunier,A. St-Arnaud,and J. Rousseau, “Monocular 3D Head Tracking to Detect Falls of Elderly People,”in 28th Annual International Conference of the IEEEEngineering in Medicine and Biology Society, pp. 6384-6387, 2006.
[10] C. Rougier, J. Meunier, A. St-Arnaud, and J. Rousseau, “Robust Video Surveillance for Fall Detection Based on Human Shape Deformation,” in IEEE Transactions on Circuits and Systems for Video Technology, vol. 21, no. 5, pp. 611-622, 2011.
[11] S. G. Miaou, P. H.Sung,and C. Y.Huang, “A Customized Human Fall Detection System Using Omni-Camera Images and Personal Information,”in1st Transdisciplinary Conference on Distributed Diagnosis and Home Healthcare,pp. 39-42, 2006.
[12] E. Auvinet, F. Multon, A. Saint-Arnaud, J. Rousseau, and J. Meunier, “Fall Detection With Multiple Cameras: An Occlusion-Resistant Method Based on 3-D Silhouette Vertical Distribution,” IEEE Transactions on Information Technology in Biomedicine, vol. 15, no. 2, pp. 290-300, 2011.

[13] G. Mastorakis and D. Makris, “Fall detection system using Kinect’s infrared sensor,” Real-TimeImageProcessing,vo1.9,no.4,pp.635-646,2014.
[14] R. Planinc and M. Kampel, “Introducing the use of depth data for fall detection,” Personal and Ubiquitous Computing, vol. 17, no. 6, pp. 1063-1072, 2013.
[15] A. N. Belbachir, A. Nowakowska, S. Schraml, G. Wiesmann,and R. Sablatnig, “Event-driven feature analysis in a 4D spatiotemporal representation for ambient assisted living,”in 13thInternational Conference on Computer Vision,pp. 1570-1577, 2011.
[16] M. Yu, Y. Yu, A. Rhuma, S. M. R. Naqvi, L. Wang, and J. A. Chambers, “An Online One Class Support Vector Machine-Based Person-Specific Fall Detection System for Monitoring an Elderly Individual in a Room Environment,” Biomedical and Health Informatics,vol. 17, no.6, pp. 1002-1014, 2013.
[17] M.Kepski, B. Kwolek, and I. Austvoll, “Fuzzy Inference-Based Reliable Fall Detection Using Kinect and Accelerometer,” in InternationalConferenceonArtificialIntelligenceandSoftComputing,pp.266-273,2012.
[18] Wikipedia-Hopfield network. [Online]. Available:
https://en.wikipedia.org/wiki/Hopfield_network. [Accessed: 08-Jun-2016].
[19] 類神經網路簡介. [Online]. Available:
http://www.mantraco.com.tw/tao/2003/D230705.htm. [Accessed: 09-Jun-2016].
[20] 蘇木春、張孝德編著, 機器學習：類神經網路、模糊系統以及基因演算法則，第二版,全華科技圖書，民國一百零一年。
[21] Wikipedia-Radial basis function. [Online]. Available:
https://en.wikipedia.org/wiki/Radial_basis_function. [Accessed: 08-Jun-2016].
[22] Wikipedia-Support Vector machine. [Online]. Available:
https://en.wikipedia.org/wiki/Support_vector_machine. [Accessed: 08-Jun-2016].
[23] C. W. Hsu andC. J. Lin, “A Comparison of Methods for Multiclass Support Vector Machines,” IEEE Transaction on Neural Networks, vol. 13, no. 2, pp. 415-425, 2002.
[24] 陳盈秀，「SVM類神經網路於單調性資料探勘之研究」，成功大學工業與資訊管理學系專班，2009。
[25] 深度學習-人工智能的現在與未來. [Online]. Available:
http://pansci.asia/archives/56816. [Accessed: 14-Jun-2016].
[26] Nicola Jones, “The Learning Machines,” Nature, vol. 505, pp.146-148, Jan. 2014.
[27] Y. Bengio, “Learning Deep Architectures for AI,” in Foundations and Trends in Machine Learning, vol. 2, no. 1, pp 1-127, 2009.
[28] Y. LeCun, Y. Bengio, and G. E. Hinton, “Deep learning,” Nature,vol. 521, no. 7553, pp.436-444, 2015.
[29] J. Schmidhuber, “Deep learning in neural networks: An overview,” Neural Networks, vol. 61, pp.85-117, 2015.
[30] Wikipedia -convolution neural network. [Online]. Available:
https://en.wikipedia.org/wiki/Convolutional_neural_network.
[Accessed: 28-Jun-2016].
[31] G. E. Hinton,” A Practical Guide to Training Restricted Boltzmann Machines,” in Neural Networks: Ticks of the Trade, Second Edition, Springer, pp. 599-619, 2012.
[32] M. Á. Carreira-Perpiñánand G. Hinton, “On Contrastive Divergence Learning,” Artificial Intelligence and Statistics, vol. 10,pp. 33-40,2005.
[33] Wikipedia-restricted Boltzmann machine. [Online]. Available:
https://en.wikipedia.org/wiki/Restricted_Boltzmann_machine. [Accessed: 28-Jun-2016].
[34]Deep belief networks.[Online].Available:
http://www.scholarpedia.org/article/Deep_belief_networks.[Accessed: 28-Jun-2016].
[35] Wikipedia-Standard Deviation. [Online]. Available:
https://en.wikipedia.org/wiki/Standard_deviation. [Accessed: 09-Jun-2016].
[36] Wikipedia-Principal component analysis. [Online]. Available:
https://en.wikipedia.org/wiki/Principal_component_analysis. [Accessed: 09-Jun-2016].
[37] Sensitivity and Specificity. [Online]. Available:
http://www.med.uottawa.ca/sim/data/Sensitivity_e.htm. [Accessed: 09-Jun-2016].
[38] Wikipedia-Cohen’s kappa. [Online]. Available:
https://en.wikipedia.org/wiki/Cohen%27s_kappa. [Accessed: 09-Jun-2016].
[39] 崔懷芝，「量表信度的測量：Kappa統計量之簡介」，中國醫藥大學生物統計研究所，2014。
[40]交叉驗證(cross validation). [Online].Available:
https://cg2010studio.com/2012/10/22/%E4%BA%A4%E5%8F%89%E9%A9%97%E8%AD%89-cross-validation/. [Accessed: 09-Jun-2016].
[41] Neuroph Studio. [Online]. Available:
http://neuroph.sourceforge.net/. [Accessed: 10-Jun-2016].
[42]tensorflow.[Online].Avaliable:
https://www.tensorflow.org/versions/r0.8/tutorials/mnist/tf/index.html.[Accessed: 28-Jun-2016].
[43] Libsvm. [Online]. Available:
https://www.csie.ntu.edu.tw/~cjlin/libsvm/. [Accessed: 10-Jun-2016].
[44] RBFN Matlab. [Online]. Available:
http://www.mathworks.com/help/nnet/ref/newrb.html. [Accessed: 10-Jun-2016].
[45] nntool Matlab. [Online]. Available:
http://www.mathworks.com/help/nnet/ref/nntool.html. [Accessed: 10-Jun-2016].
[46] S. Hochreiter and J. Schmidhuber, “Long Short-Term Memory,” Neural Computation, vol. 9, no. 8, pp. 1735-1780, 1997.
[47]F. A. Gers, N. N. Schraudolph, and J. Schmidhuber, “Learning precise timing with lstm recurrent networks,” The Journal of MachineLearning Research, vol. 3, pp. 115-143, 2003.

簡易檢索 / 詳目顯示

相關論文